Annex III: EBA DPM translation

The content of EBA’s DPM is imported into the SMCube methodology automatically from the XBRL taxonomies. All the DPM content remains untouched, only a translation between methodologies is performed. All the imported items belong to the maintenance agency “European Banking Authoritiy (EBA)”, reflecting the fact that the content is left exactly as published by the EBA through the taxonomies.

Core package translation

The core package of the SMCube methodology is very similar to the dictionary layer of the DPM, so in most cases the translation is very straightforward.
The most relevant changes that should be noted are:

  • Metrics are members of the domain Amount type in the DPM. As explained above, in SMCube metrics are treated as variables. Nevertheless, metrics are also imported as members, because the DPM contains member hierarchies with in which the members are metrics, so they get imported in the SDD as members in order not to lose the hierarchies.
  • New domains are added with the metric data types. The reason is the conversion of metrics into variables. Variables in SMCube are defined on a domain, and metrics have a data type, so the data types of the metrics are created as domains. These new domains are:
    • Monetary
    • String
    • Percent
    • Boolean
    • Date
    • Integer
    • Decimal
  • New variables are added to make explicit some dimensions that the DPM retains implicit, but that are included in the XBRL instances. These variables deal with the reporting agent, the date and the observation value, and are:
    • Period
    • Entity
    • Unit
    • Observation_Value

Data definition package translation

The translation of the data definition package is not straightforward, since the DPM does not represent multidimensional structures. Therefore, a convention to what is equivalent to a cube is required. The convention followed has been translating one table as one cube.
In the cube structure items, the implicit variables are added to the cube. All the DPM dimensions have the dimension role also under the SMCube methodology.
Data points are translated as combinations: each combination within a non-reference cube is considered a data point described by the DPM. The FinRep reference cubes are a normalised template described by reference codes and reference combinations.
It is possible to link a non-reference combination – a datapoint – to a reference combination1.


  1. The non-reference combinations ID are coded using DataPointVID of DPM database. Reference combinations ID are structured in the same way, ending with a “_REF” string. (i.e. EBA_100 <-> EBA_100_REF, and identify DataPointVID = 100).